The alternative liquid crystal display technique, using passive rather than active eyewear, is one championed by Arisawa. They make a so-called “Xpol” material that is in fact a sheet of micro-retarder overlaid on an LC screen so that alternate lines of retardation line up with rows of pixels. Because liquid crystal displays depend on polarization for image formation, the polarized light that emerges from the display is phase-shifted by the micro-retarder as noted in alternate lines. It’s an interlace-type display, so we get alternate lines of left- and right-handed circularly polarized light that can be analyzed with circular polarizing glasses similar, or possibly identical, to those used in the stereoscopic cinema.
Polarizing eyewear are a commodity product. The reason I am pointing out that both the shuttering eyewear and the passive polarizing eyewear are or will become commodity products is that the people who are going to make out financially from 3D TV is probably may not be inventors and innovators or owners of intellectual property but rather a factory in China or Taiwan. Although it may seem unlikely to the reader, someday shuttering eyewear could be virtually as inexpensive as passive circular polarizing eyewear. For example, a plastic liquid crystal shutter would change the fabrication technique and change the entire cost of goods equation. Such devices have been demonstrated and can be manufactured on web machines.
Thus we will have two types of 3D liquid crystal TV sets: One that has a fast refresh rate, say 120 fields per second, and uses active or shuttering eyewear, and another that uses alternative lines of polarization encoding and polarizing passive eyewear. If we take a look at the television set itself and see who is going to make out in a financial sense, it’s obviously the television set manufacturer who is going to be selling television sets. If the television set uses a fast refresh rate, the manufacturing technique involved is not going to add cost of goods. That’s because television set manufacturers, as noted, have been trying to make their liquid crystal displays look like CRT displays, so they have made a major effort in terms of speeding up their refresh rate in order to make motion look crisper and sharper. Pans, rapid motion, those things, look better at a high refresh rate, so 120Hz sets and even 240Hrz sets have been coming online for the past couple of years.
There is no additional cost of goods for such sets to be stereoscopic, which means that the television set manufacturer can sell such a set to the public for no additional cost of goods, except for the fact that there will have to be a stereoscopic version of the scaling engine – a stereo engine. We’ll talk more about the stereo engine in a moment. The public would then be required to buy shuttering eyewear. The set might have an emitter built into it, or the emitter might be an auxiliary add-on. This is more additional cost that may or may not be directly associated with the television set manufacturer, but there would be a burden on the public to buy these accessories that would allow the customer to see stereoscopically. But of course it’s always a choice you could make or not make. In the early days the glasses are likely to be more expensive than in long run, when economies of scale set in.
With regard to the Xpol or Micropol sets, those are going to cost more because an additional layer of some kind – a retardation layer – has to be placed on the surface of the display. And they will require a stereo scaling chip to. There is nothing proprietary about the idea of polarizing pixels or television lines. Over a century ago people thought of the idea. What is proprietary, and what is key here, is the manufacturing technique. It’s pretty obvious that you need a web process to manufacture sheets of this stuff, and tooling up for a web process would be quite costly. I don’t know at this moment whether or not Arisawa, probably the leading contender in the field, has actually done so, but allegedly there is a JVC monitor coming out pretty soon that will use Arisawa material, and one assumes they have committed to the web line.
Stereoscopic television sets are going to require a stereo engine because the signals received by the set will be many and varied, and it will depend on the distribution system. By “distribution system” I mean disc, IPTV, terrestrial broadcast, cable, cable on demand, or whatever. Each distribution system will have its own protocol or format to suit its own needs. It won’t necessarily have to be transparently downwardly compatible. There is probably just no need for this characteristic in this day and age, for reasons I’ve given earlier. But it does mean that there will not be a dominant muxing signal. All of these muxing formats will have to live with existing MPEG or JPEG format requirements, so that whatever stereoscopic formats somebody dreams up, the information can’t be destroyed with the fundamental multiplexing technique.
I submit that because of Philip’s unfortunate withdrawal, in a practical sense, we are no longer concerned with autostereo. I know there are some technologists at the studios in particular who say that if you’re going to create a standard for stereoscopic television now, you ought to create autostereo standards as well so we don’t have to go back and do it again. But that would be like asking the people who defined NTSC television to also create standards for digital and hi-def TV. We just don’t know what’s coming down the pike, so we’re going to have to live with what may turn out to become an interim solution – but that interim could be decades.